Survey data show that Mexican-American (MA) youth report the highest rates of experimenting with cigarettes. This transdisciplinary project will determine the relative influences of both non-genetic (psychosocial, behavioral, and contextual) and genetic factors in susceptibility to smoking, initiation of smoking, and dependency on nicotine in a cohort of urban MA youth in the Houston metropolitan area. The proposal builds upon an existing population-based resource of MA households created and maintained in the Department of Epidemiology at M. D. Anderson Cancer Center, and capitalizes upon the experience in studying adolescent smoking trajectories in the Department of Behavioral Science. The study design will include an initial cross-sectional survey with the following specific aims: 1) to assess baseline prevalence of susceptibility to smoking and experimentation with cigarettes in the prior six months in unrelated MA adolescents between the ages of 11 to 13 years (n =1300). 2) to conduct a systematic telephone follow-up every six months to ascertain changes in smoking status (using widely accepted measures of cognitive susceptibility to smoking and nicotine dependence) and to collect saliva samples (as needed) for cotinine measurement. We will evaluate changes in psychosocial and contextual factors by conducting a follow-up home interview in the final year of the study. The hypothesis is that adolescents with lower levels of psychosocial resources or in smoking permissive contexts are more susceptible to smoking and to the development of dependent smoking. 3) to determine how variation in candidate genes, including those related to nicotine metabolism, dopamine and serotonin pathways, and genes encoding enzymes involved in synthesis or metabolism of neurotransmitters may influence the initiation, establishment of smoking, and nicotine dependence following experimentation. We will explore how genetic predisposing factors interact with psychosocial factors to impact smoking status. We will also evaluate whether genetic predisposition will modify the impact of the social, contextual, and psychological factors on the transition from experimentation to nicotine dependence. The long-term goal is to develop quantitative multivariate risk assessment models for experimentation, initiation and dependence, and genetic susceptibility in order to identify high-risk adolescent subgroups. These findings will enable us to develop culturally and age-appropriate school- and community-based smoking cessation interventions.